Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A computer-automated method for segmenting image data for an organ of a subject, where the organ is a tube, comprising: receiving, by an image processor, image data representing a volume of the subject, such that the image data includes the organ; generating, by the image processor, a centerline through of the organ; determining, by the image processor, location of an outer wall of the tube within the image data, where the location of the outer wall is determined using the centerline; determining, by the image processor, location of an inner wall of the tube within the image data, where the location of the inner wall is determined using the outer wall; and computing, by the image processor, a measure of the organ from the image data; wherein the location of an outer wall or the location of an inner wall is determined by a) identifying a first seed point within the image data, where the first seed point is located on the centerline; b) generating a plurality of inner wall line segments that extend radially outward from the first seed point and that are orthogonal to the centerline, where each line segment in the plurality of inner wall line segments terminates at an intersection with a wall; c) identifying a magnitude of brightness of a pixel of the first seed point; d) identifying a magnitude of brightness of an additional pixel located along a length of each line segment in the plurality of inner wall line segments, where the additional pixel is moving radially outward from the first seed point; and e) repeating step d) until the magnitude of brightness of the additional pixel located along each line segment in the plurality of inner wall line segments is greater than a predetermined amount.
This invention relates to automated image segmentation of tubular organs in medical imaging. The method addresses the challenge of accurately identifying and measuring the walls of tubular structures, such as blood vessels or intestines, within volumetric image data. The process begins by receiving image data containing the organ of interest. An image processor generates a centerline through the tubular organ, which serves as a reference for subsequent segmentation steps. The outer wall of the tube is then located using the centerline, followed by the inner wall, which is determined based on the outer wall's position. The method computes a measurement of the organ, such as wall thickness or lumen diameter, from the segmented data. The segmentation involves selecting a seed point on the centerline and generating radial line segments orthogonal to the centerline. Each segment extends outward until encountering a wall, identified by a brightness threshold. The brightness of pixels along each segment is compared to a predetermined value to determine wall intersections. This approach ensures precise segmentation by leveraging intensity gradients and geometric constraints. The technique is particularly useful in medical imaging for diagnosing conditions related to tubular organ morphology.
2. The method of claim 1 further comprises capturing the image data using one of computer topography or magnetic resonance imaging.
3. The method of claim 1 wherein generating the centerline further comprises: a) identifying an origin point within the image data, where the origin point is located at one end of the tube; b) generating a plurality of line segments that extend radially outward from the origin point and into the tube, each line segment in the plurality of line segments terminates at an intersection with the inner wall of the tube; c) identifying a given line segment having longest linear dimension amongst the plurality of line segments; d) generating a plurality of additional line segments that extend radially outward from a terminus of the given line segment, where the terminus is located at the intersection with the inner wall, and each line segment in the plurality of additional line segments terminates at the intersection with the inner wall; and e) repeating steps c) and d) until the given line segment extends outside of the tube.
This invention relates to a method for generating a centerline within a tubular structure, such as a pipe or conduit, using image data. The method addresses the challenge of accurately determining the central path of a tube, which is essential for applications like inspection, maintenance, or navigation within confined spaces. The process begins by identifying an origin point at one end of the tube in the image data. From this origin, multiple line segments are generated, extending radially outward into the tube and terminating at intersections with the inner wall. The longest line segment among these is selected, and additional line segments are generated from its endpoint, again terminating at the inner wall. This selection and extension process repeats iteratively, with each new longest segment serving as the starting point for further extensions, until the generated line segments extend beyond the tube. The resulting path forms the centerline, representing the central axis of the tubular structure. This approach ensures precise tracking of the tube's center, even in complex geometries, by dynamically adapting to the tube's curvature and ensuring continuous progression toward the opposite end.
4. The method of claim 3 further comprises joining the given line segment to form a spline curve.
A method for constructing a spline curve from a given line segment involves generating a spline curve by joining the line segment with additional segments. The method first generates a spline curve from a set of control points, where the control points define the shape of the curve. The spline curve is constructed by interpolating between the control points to form a smooth, continuous curve. The given line segment is then joined to this spline curve, extending or modifying the curve's shape. The joining process ensures that the line segment connects seamlessly with the spline curve, maintaining smoothness and continuity. This method is useful in computer-aided design, animation, and other applications where precise curve generation is required. The technique allows for flexible curve construction by combining predefined line segments with dynamically generated spline curves, improving design accuracy and efficiency.
5. The method of claim 3 wherein the origin point is located substantially near a center of a cross section of the tube.
6. The method of claim 1 wherein computing the measure of the organ further comprises at least one of a thickness between the inner wall and the outer wall, diameter of the tube, diameter of the outer wall, length of a diseased section, and contrast assessment.
7. The method of claim 1 wherein determining the location of the outer wall of the tube further comprises: creating a planar reformation of the volume using the centerline; for each slice of the volume, converting data for a given slice from a cartesian coordinate to a polar coordinates by generating a plurality of wall line segments that extend radially outward from the first seed point and that are orthogonal to the centerline; identify candidates that exceed the outer wall magnitude threshold and iteratively add or remove candidates that are proximal to a best fit grid; transform grid candidates from the polar coordinates to the Cartesian coordinates; and remap the transformed grid points to an original coordinates.
This invention relates to medical imaging, specifically to methods for accurately determining the location of the outer wall of a tubular structure, such as a blood vessel, in volumetric imaging data. The problem addressed is the difficulty in precisely identifying the boundaries of such structures due to noise, artifacts, and variations in image quality, which can lead to inaccuracies in clinical assessments and diagnostic measurements. The method involves creating a planar reformation of the volumetric data using a centerline of the tubular structure. For each cross-sectional slice of the volume, the data is converted from Cartesian to polar coordinates. This conversion involves generating multiple wall line segments that extend radially outward from a predefined seed point and are orthogonal to the centerline. The method then identifies potential wall candidates that exceed a predefined magnitude threshold and iteratively refines these candidates by adding or removing points that are proximal to a best-fit grid. The refined grid candidates are then transformed back from polar to Cartesian coordinates and remapped to the original coordinate system. This approach improves the accuracy of outer wall detection by leveraging geometric transformations and iterative refinement techniques. The method is particularly useful in applications requiring precise measurements of tubular structures, such as vascular imaging and cardiovascular diagnostics.
8. The method of claim 1 wherein the predetermined amount is two standard deviations of the magnitude of brightness of the pixel of the first seed point.
9. The method of claim 1 wherein the organ is further defined as a small intestine and an origin point is located at an intersection between the small intestine and a large intestine of the subject.
This invention relates to medical imaging and surgical navigation, specifically for identifying anatomical landmarks in the gastrointestinal tract. The method addresses the challenge of precisely locating anatomical features within the small intestine, which is critical for minimally invasive surgical procedures or diagnostic imaging. The small intestine is a highly mobile and convoluted organ, making it difficult to consistently identify specific regions, such as the intersection between the small intestine and the large intestine. This intersection serves as a critical reference point for surgical planning, biopsy targeting, or therapeutic interventions. The method involves using imaging data, such as computed tomography (CT) or magnetic resonance imaging (MRI), to analyze the gastrointestinal anatomy. The system processes the imaging data to detect and highlight the intersection between the small intestine and the large intestine, which is defined as the origin point. This origin point is used as a reference for further navigation or intervention within the small intestine. The method may also incorporate real-time tracking during surgery to ensure accurate alignment with the preoperatively identified origin point. By providing a consistent and reliable reference, the technique improves the precision of surgical procedures and reduces the risk of complications associated with incorrect anatomical targeting. The approach is particularly useful in robotic-assisted surgeries or endoscopic interventions where precise navigation is essential.
10. A method for segmenting an organ of a subject, comprising: a) receiving image data of the organ of the subject, wherein the image data includes at least a small intestine of the subject; b) identifying an origin point within the image data, wherein the origin point is located at an intersection between the small intestine and a large intestine of the subject; c) generating a plurality of line segments that extend radially outward from the origin point, each line segment in the plurality of line segments terminates at an intersection with a first boundary of the organ; d) identifying a given line segment having longest linear dimension amongst the plurality of line segments; e) identifying a point along the given line segment to serve as a seed location; f) generating a plurality of additional line segments that extend radially outward from the seed location, each line segment in the plurality of additional line segments terminates at the intersection with the first boundary; and g) repeating steps d) f) until the given line segment extends outside of the organ.
11. The method of claim 10 wherein generating the plurality of line segments and generating the plurality of additional line segments includes using a random walker analysis.
12. The method of claim 10 further comprises capturing the image data using one of computed tomography or magnetic resonance imaging.
This invention relates to medical imaging techniques, specifically methods for capturing and processing image data to enhance diagnostic accuracy. The method involves acquiring image data of a patient's anatomy using either computed tomography (CT) or magnetic resonance imaging (MRI). The captured image data is then processed to generate a three-dimensional (3D) representation of the anatomy, which is used to identify and analyze specific regions of interest within the patient's body. The method further includes segmenting the 3D representation to isolate these regions, allowing for detailed examination of anatomical structures. Additionally, the method may involve applying machine learning algorithms to the segmented data to detect abnormalities or pathologies, such as tumors, fractures, or other medical conditions. The processed data can be visualized in various formats, including cross-sectional views, 3D renderings, or augmented reality overlays, to assist healthcare professionals in diagnosis and treatment planning. The use of CT or MRI ensures high-resolution imaging, enabling precise identification and analysis of anatomical features. This approach improves diagnostic accuracy and supports personalized medical interventions.
13. The method of claim 10 further comprises generating a centerline through the organ by joining the given line segment.
14. The method of claim 13 further comprises determining a location of the first boundary, wherein determining the location includes: a) identifying a first seed point within the image data, where the first seed point is located on the centerline; b) generating a plurality of first boundary line segments that extend radially outward from the first seed point and that are orthogonal to the centerline, where each line segment in the plurality of first boundary line segments terminates at an intersection with the first boundary; c) identifying a magnitude of brightness of a pixel of the first seed point; d) identifying a magnitude of brightness of an additional pixel located along a length of each line segment in the plurality of first boundary line segments, where the additional pixel is moving radially outward from the first seed point; and e) repeating step d) until the magnitude of brightness of the additional pixel located along each line segment in the plurality of first boundary line segments is greater than a predetermined amount.
15. The method of claim 14 further comprises determining a location of a second boundary, wherein determining the location includes: a) identifying a second seed point within the image data, where the second seed point is located on the first boundary; b) generating a plurality of second boundary line segments that extend radially outward from the second seed point and that are orthogonal to the first boundary, where each line segment in the plurality of second boundary line segments terminates at an intersection with the second boundary; c) identifying a magnitude of brightness of a pixel of the second seed point; d) identifying a magnitude of brightness of an additional pixel located along a length of each line segment in the plurality of second boundary line segments, where the additional pixel is moving radially outward from the second seed point; and e) repeating step d) until the magnitude of brightness of the additional pixel located along each line segment in the plurality of second boundary line segments is greater than the predetermined amount.
This invention relates to image processing techniques for boundary detection within image data. The method addresses the challenge of accurately identifying and delineating boundaries in images, particularly where brightness variations complicate boundary detection. The process involves determining the location of a second boundary by first identifying a seed point on an already-determined first boundary. From this seed point, multiple line segments are generated, extending radially outward and orthogonal to the first boundary. Each segment terminates where it intersects the second boundary. The method then analyzes brightness values along these segments, starting from the seed point and moving outward. For each segment, the brightness of pixels is compared to a predetermined threshold. The process continues until the brightness of a pixel along any segment exceeds this threshold, indicating the presence of the second boundary. This approach ensures precise boundary detection by leveraging brightness gradients and orthogonal projections from an initial boundary. The technique is useful in applications requiring accurate segmentation, such as medical imaging, object recognition, and automated image analysis.
16. The method of claim 15 further comprises: generating a three-dimensional point cloud of the organ based on the centerline, the first boundary, and the second boundary; and measuring a metric of the organ of the subject.
17. The method of claim 16 wherein the first boundary corresponds to an inner wall of the organ and the second boundary corresponds to an outer wall of the organ.
18. The method of claim 17 wherein the metric of the organ includes at least one of a thickness between the inner wall and an outer wall, diameter of the inner wall, diameter of the outer wall, length of a diseased section, and contrast assessment.
This invention relates to medical imaging and analysis, specifically for assessing organ health by measuring specific anatomical and functional metrics. The method involves analyzing an organ, such as a blood vessel, to determine key structural and functional characteristics. These metrics include the thickness between the inner and outer walls of the organ, the diameter of the inner wall, the diameter of the outer wall, the length of a diseased or affected section, and contrast assessment to evaluate blood flow or tissue perfusion. The technique likely involves imaging modalities such as ultrasound, CT, or MRI, combined with image processing algorithms to extract these measurements. The goal is to provide quantitative data for diagnosing conditions like atherosclerosis, aneurysms, or other vascular diseases, enabling early detection and treatment planning. The method may also include comparing these metrics against reference values to assess disease severity or progression. By automating the measurement process, the invention aims to improve diagnostic accuracy, reduce human error, and streamline clinical workflows. The contrast assessment aspect may involve analyzing the distribution and intensity of contrast agents within the organ to evaluate blood flow dynamics or tissue viability. This approach enhances the diagnostic capabilities of medical imaging systems, supporting more precise and personalized patient care.
19. A computer-automated method for segmenting image data for an organ of a subject, where the organ is a tube, comprising: receiving, by an image processor, image data representing a volume of the subject, such that the image data includes the organ; generating, by the image processor, a centerline through of the organ; determining, by the image processor, location of an outer wall of the tube within the image data, where the location of the outer wall is determined using the centerline; determining, by the image processor, location of an inner wall of the tube within the image data, where the location of the inner wall is determined using the outer wall; and computing, by the image processor, a measure of the organ from the image data; wherein the centerline is generated by a) identifying an origin point within the image data, where the origin point is located at one end of the tube; b) generating a plurality of line segments that extend radially outward from the origin point and into the tube, each line segment in the plurality of line segments terminates at an intersection with the inner wall of the tube; c) identifying a given line segment having longest linear dimension amongst the plurality of line segments; d) generating a plurality of additional line segments that extend radially outward from a terminus of the given line segment, where the terminus is located at the intersection with the inner wall, and each line segment in the plurality of additional line segments terminates at the intersection with the inner wall; and e) repeating steps c) and d) until the given line segment extends outside of the tube.
Unknown
February 16, 2021
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.